KR100823467B1 - System and method to provide fairness and service differentiation in ad-hoc networks - Google Patents

Abstract

The present invention is directed to a system and method for a medium access control (MAC) scheduling algorithm having a distributed self-negotiation method for providing usage fairness and service discrimination in an ad hoc network by using control RTS / CTS messages carrying node state information. It is about. The local scheduling algorithm uses overheard RTS / CTS information with attributes that specify node state to maximize the awareness of the neighbor state for a multi-channel system, where the data and ACK messages are different data channels. Is sent through the network. If the system uses a single channel, the corresponding information can be carried by using DATA or ACK messages. The algorithm can be used to adjust the channel access timer to provide fairness of use between different nodes, different links, and different streams, such as nearby offered loads, carried loads, and backlogged loads. Measure The algorithm calculates a priority level based on node queue state and a priority level based on neighbor state to destroy contention tiers and enable service differentiation.

Ad hoc network, MAC, RTS, CTS, scheduling, priority level weights

Description

System and method to provide fairness and service differentiation in ad-hoc networks}

This application discloses 35 U.S. Claims US Preliminary Patent Application 60 / 476,168, filed June 6, 2003, under § 119 (e), the entire contents of which are incorporated herein by reference.

The present invention relates to a medium access control (MAC) protocol in a wireless ad hoc network system. In particular, communication systems and methods with handshaking protocols are used for data exchange in single or multichannel, multihop wireless networks that include local scheduling algorithms to solve fairness and service differentiation problems in a fully distributed system. do.

Wireless communication networks, such as mobile wireless telephone networks, have become widespread over the last few decades. These wireless communication networks are referred to as "cell networks" because the network infrastructure is arranged to divide the service area into multiple regions called "cells." A terrestrial cellular network includes a number of interconnected base stations, or base nodes, geographically distributed at locations designed through a service area. Each base node includes one or more transceivers capable of transmitting and receiving electromagnetic signals, such as radio frequency (RF) communication signals, to and from mobile user nodes, such as wireless telephones disposed within a coverage area. Communication signals include, for example, voice that is modulated according to the desired modulation technique and transmitted as a data packet. As will be appreciated by those skilled in the art, network nodes may be configured in a time division access (TDMA) format, code division multiple access (CDMA) format, or frequency division multiplexing such that a single transceiver at a base node communicates with several mobile nodes simultaneously in a coverage area. Send and receive data packet communications in a multiplex format, such as an access (FDMA) format.

Recently, one type of mobile communication network called "ad-hoc" network has been developed for military use. In this type of network, each mobile node can act as a base station or as a router for other mobile nodes, thus eliminating the need for a fixed infrastructure of base stations. Details of the ad hoc network are described in US Pat. No. 5,943,322 by Mayor, the entire contents of which are incorporated herein by reference.

More sophisticated ad hoc networks, in addition to mobile nodes being able to communicate with each other as in conventional ad hoc networks, in addition, mobile nodes access fixed networks and other mobile nodes such as public switched telephone networks (PSTNs), and the Internet It is being developed to communicate with other networks. Details of these advanced forms of ad hoc networks are described in US patent application 09 / 897,790, filed June 29, 2001, entitled "Ad Hoc Peer-to-Peer Mobile Radio Access System Interfaced to the PSTN and Cellular Networks." , US Patent Application No. 09 /, filed March 22, 2001, entitled " Time Division Protocol for an Ad-Hoc, Peer-to-Peer Radio Network Having Coordinating Channel Access to Shared Parallel Data Channels with Separate Reservation Channel. &Quot; 815,157, and US Patent Application 09 / 815,164, filed March 22, 2001, entitled " Prioritized-Routing for an Ad-Hoc, Peet-to-Peer, Mobile Radio Access System " The entire contents are incorporated herein by reference.

However, communication between nodes is subject to errors due to interference, multipath and fading effects and collisions. Avoidance of many of these errors can be achieved using control signal handshake between transmitting and receiving nodes. Communication protocols such as multiple access (MACA) with collision avoidance consist of request-to-send (RTS) control sent from a source node to a destination node that responds to the source node with a clear-to-send (CTS) control packet. It uses handshake techniques between nodes. However, Multiple Access with Collision Avoidance (MAKA) algorithms typically handle ARQ retransmissions for correction of errors by repeating the entire RTS / CTS channel access sequence. In addition, MACAW introduced Data-Sending (DS) messages to form the RST-CTS-DS-DATA-ACK message exchange and a new backoff algorithm. IEEE 802.11 MAC is a multiplicity of CSMA / CA protocols that implement both carrier sensing and virtual (RTS-CTS switched) carriers that detect acknowledgment (hereinafter referred to as "Ack") messages to improve reliability.

Thus, there is a need for a media access control protocol that detects and schedules communications in a wireless ad hoc network system to avoid the errors mentioned above and to eliminate the complexity associated with modification.

A method for avoiding multiple access collisions in a MACA based wireless network is described, as described in US Pat. No. 6,556,582 to Redi, the entire contents of which are incorporated herein by reference. This collision avoidance method has been proposed in systems with multiple transceivers, namely data channel and reserved channel transceivers. Since the transceivers operate simultaneously, the reservation channel transmission is not omitted. NCTS messages are used to inform other nodes that the intended destination node is busy. Thus, it is necessary to solve the problems associated with having one transceiver and one or multiple data channels, such as omitted reservation data. The method described in the Redi patent involves using a priority field for RTS messages. However, the processing of this information is based on absolute comparisons, and the relative state of transmission between neighbors is not used and fairness is not considered.

As discussed in US Pat. No. 6,118,788 to Kermani et al., Which is incorporated herein by reference in its entirety, a method for providing fairness in a MACA based wireless network is described. The system considered in the Kermani patent does not have transmission durations distributed in RTS-CTS messages, but only one channel that is distributed through the end of the bus control frame after transmission. Thus, there is a need for address and channel monitoring involving multiple data channel systems, where transmission lengths are advertised via RTS-CTS messages. The main objective in the Kermni patent is to solve hidden terminal problems, not to solve problems associated with having one transceiver and multiple data channels, such as omitted reservation data. The system proposed in the Kermani patent is based on the fact that by distributing this information, the neighbors will have sufficient knowledge of the number or some other information of possible and logical connections in the neighborhood. For example, it is proposed that neighbors listening to the advertised window size will adjust accordingly. However, if the information is incorrect, this can cause the problem that all nodes reduce the backoff time and increase the collision. Thus, there is a need for a method and system that uses feedback from transmissions and receiver assistant information correction to complete information and links about neighbors. Service differentiation is not disclosed in the Kermani patent.

As discussed in US Patent Application Publication No. 20020154653 by Benveniste, the contents of which are incorporated herein by reference, a method of applying a backoff to a track for a CSMA network is described. The system considered is one hop system with one channel. The method aims to improve IEEE 802.11 type networks in terms of contention delay and service differentiation. However, in multi-hopping systems, one important cause of channel access delay is that many streams can be collected at one relay node (such as wireless routers that do not centrally control capacity as access points in one hop systems) and relay The node can be busy. Therefore, although channel and address monitoring reduces the likelihood of collisions (and therefore reduces the number of faulty transmission attempts), the traffic density provided can be high. This delay can cause short term inequalities and can slow TCP type traffic, changing the actual traffic density. Therefore, node weighting is needed to alleviate this problem.

In addition, there is a need to differentiate reserved channel and data channel collisions for systems with multiple channels. The other point is that, unlike the method described in the Benveniste publication, streams for other next hops exist in the relay nodes and backlog packets due to next hop busy state or link failure may be scheduled later than newer packets scheduled for other next hops. Can be. Incomplete inter-neighbor information is a major problem in multihop, one or multi-channel networks, and therefore, local measurement (via unicast messaging) comparison systems and methods are needed to alleviate these problems.

It is an object of the present invention to provide a medium access control (MAC) system and method comprising a local scheduling algorithm, in particular bidirectional scheduling to provide fairness and service differentiation in an ad hoc network using RTS / CTS control messages carrying node state information. (two-way scheduing) technology.

Another object of the present invention is to provide a local scheduling algorithm that can measure single node information using RTS / CTS message information and assigns a weight factor to a single node based on the measured information.

Another object of the present invention is to provide a local scheduling algorithm for measuring neighbor node information of a single node using RTS / CTS message information and assigning a weight factor to each neighbor node of the single node based on the measured information. .

It is another object of the present invention to include a first stateless method using a weight factor of a single node, and a second stateful method using a weight factor of a single node and a weight factor of a neighbor node. It is to provide a local scheduling algorithm capable of executing at least two scheduling methods.

Another object of the present invention is to provide a local scheduling algorithm capable of executing at least two scheduling methods for regulating channel access to provide fair use among nodes, other links and other streams.

These and other objects are achieved by providing a MAC algorithm system and method with a distributed self-regulation method to provide usage fairness and service differentiation in an ad hoc network by using control RTS / CTS messages that carry node state information. The local scheduling algorithm is used to provide eavesdropping RTS / CTS information with attributes representing the state of the node to maximize the awareness of the neighbor. The system may consist of a single channel or multiple channels in which data and ACK messages are transmitted over other channels. The algorithm measures the provided load, carried load, backlogged load in the neighborhood to adjust the channel access timer to provide usage fairness among other nodes, other links, and other streams. The algorithm may calculate a priority level based on the node queue state, and also calculate the priority level based on the neighbor state to destroy the tiers and differentiate the service.

These and other objects, advantages and novel features of the present invention will become more readily apparent from the following detailed description when read in conjunction with the accompanying drawings.

1 is a block diagram of an exemplary ad hoc packet switching wireless communication network including a plurality of nodes in accordance with an embodiment of the present invention.

FIG. 2 is a block diagram illustrating an embodiment of a mobile node used in the network shown in FIG. 1.

In the embodiments of the present invention described below, a communication system and method is disclosed that includes a handshaking protocol for data exchange in a single / multichannel multihop wireless network. Local scheduling algorithms are provided to one or more nodes to address channel usage fairness and service differentiation issues in a fully distributed system, such as network 100 of FIG.

1 is a block diagram illustrating an embodiment of an ad hoc packet switch wireless communication network 100 embodying an embodiment of the present invention. In particular, network 100 includes a plurality of wireless user terminals 102-1 through 102-n (generally referred to as node 102 or mobile nodes 102), and in fixed network 104. In order to provide access to nodes 102, it is not required, but multiple access points 106-1, 106-2,... 106-n (referred to as nodes 106 or access points 106). It may include a fixed network 104 having a. The fixed network 104 may provide network nodes with access to other ad hoc networks, public switched telephone networks (PSTNs), and other networks, such as the Internet, for example, a core local access network (LAN), and It may include multiple servers and gateway routers. Network 100 uses a number of fixed routers 107-1 through 107-n (called nodes 107 or routers 107) to route data packets between other nodes 102, 106 or 107. It may include. For this discussion, the nodes described above may be collectively referred to as "nodes 102, 106 and 107", or simply "nodes" or "terminals." As can be appreciated by one of ordinary skill in the art, nodes 102, 106, and 107 can communicate with one or more other nodes operating directly or as routers for packets sent between nodes, which is US Pat. Nos. 5,943,322 and 09 / 897,790, 09 / 815,157 and 09 / 815,164 to Mayor, supra, referenced above.

As shown in FIG. 2, each node 102, 106 and 107 is coupled to an antenna 110 and signals packaged to and from the node 102, 106 or 107 under the control of the controller 112. And a transceiver 108 capable of receiving and transmitting the same signals. Packaged data signals include, for example, voice, data or multimedia information, package control signals including node routing and update the information.

Each node 102, 106, and 107 further includes a memory 114, such as random access memory (RAM), which may store, among other things, routing information belonging to itself and other nodes in the network. Nodes 102, 106, and 107 communicate respective routing information, called routing advertisements, to each other through a broadcast mechanism at various intervals, such as when a new node enters the network or existing nodes in the network move. Replace periodically.

As further shown in FIG. 2, certain nodes, especially mobile nodes 102, may be comprised of any number of devices, such as a notebook computer terminal, a mobile phone unit, a mobile data unit, or any other suitable device. May include a host 116. Each node 102, 106, and 107 includes suitable hardware and software to perform the Internet Protocol (IP) and Address Resolution Protocol (ARP), the purpose of which can be readily appreciated by those skilled in the art. . Hardware and software suitable for performing Transmission Control Protocol (TCP) and User Datagram Protocol (UDP) may be included. In addition, each node has Automatic Repeat Request (ARQ) functions and media access control (MAC) including a local scheduling algorithm provided to address fairness and service differentiation issues as described in more detail below. Hardware and software protocols suitable for performing < RTI ID = 0.0 >

In the network 100 of FIG. 1, each node 102, 106, and 107 may operate as a local scheduler by self-adjustment based on amounts measured in time. This functionality can be achieved through the use of a "scheduling algorithm" that can be executed in several ways based on measured quantities of loads provided and carried by balancing a tradeoff between performance and complexity. The local scheduling table is maintained at every node and updated by using address and channel monitoring results. In addition, each receiving node can control scheduling by using its own table and responding to tuning factors. Examples for tuning the factor are provided in the example algorithms. Thus, the embodiment described below solves a number of problems related to hidden and exposed nodes and problems related to movement by using a bidirectional scheduling technique initiated by the transmitter and coordinated by the receiver.

In the embodiment of the present invention described below, not only a single channel but also multiple channels may be used. Examples are provided for multichannel systems. In a multi-channel multi-hop network, such as network 100 of FIG. 1, "Request-To-Send" (RTS) and "Clear-To-Send" (CTS) messages It is sent over reservation channels, and data and "ACK" messages are sent over selected data channels. However, in a single channel system, these messages are sent over the same channel. RTS / CTS messages are weighted queue length (ie, priority, calculated from source and destination MAC addresses, current packet length, data channel rate, selected data channel, and packet priority levels). Neighbor state information, which may include retransmission attempts, packet age, etc.).

In this embodiment, at least two scheduling methods can be applied based on the information derived from the RTS / CTS messages described above. "Stateless" scheduling can be performed using only the node's own priority level, which is called the node's "weight." "Stateful" scheduling may be performed by further adjusting in accordance with the weight values of the neighboring node. By performing stateful scheduling, each node can maintain neighbor (ie neighbor node) information from the sent RTS / CTS control messages.

In the embodiment of the present invention described below, a local scheduling algorithm is provided for controlling channel access time and backoff delays by using current state information of at least one node itself, and neighbor node current state information of the node. In addition, the scheduling algorithm provides service differentiation by adjusting these timers according to priority indices.

Local scheduling algorithms are based on bidirectional distributed, self-coordinating methods to provide fairness and service differentiation in ad hoc networks by using reservation data information. Final distributed scheduling in an ad hoc network including any number of mobile nodes may be achieved through the following steps in accordance with the first and second embodiments of the present invention.

In a first step, the weighting factor of traffic load and priority is measured at each node. In a second similar step, the weighting factor of traffic load and priority at each node is measured for each neighboring node. These values are used alone or in combination with a third step to calculate at the transmitting node an access time interval based on traffic at the transmitting node, and neighboring traffic conditions of the transmitting node. This includes the first part of two-way scheduling, initiated by the transmitter and coordinated by the receiver, as described below in four and five steps.

In a fourth step of the receiving node, the same bandwidth and access time interval for each remote node is calculated based on measured traffic obtained from unintended RTS and CTS as described by way of example, and then in accordance with the calculated values. A fifth step is followed at the receiving node to decide whether to send a CTS or NCTS with adjusted channel access times (ie, a tuning factor that reduces or increases channel access attempts) based on node access. It is initiated by the transmitter and includes the second part of the two way scheduling coordinated by the receiving node.

The five steps of the bidirectional scheduling algorithm described above may be applied as the first or second embodiment in a stateless or stateful scheduling algorithm according to the targeted scheduling method. As mentioned above, two scheduling methods may be employed. Stateless scheduling may be performed using the node's own weights, and stateful scheduling may be performed by further adjusting according to neighbor node weight values.

In the first embodiment of the scheduling algorithm, only address and channel monitoring are used for information collection and transmission time scheduling after the address and / or channel is empty. In a second embodiment of the scheduling algorithm, each node collects information about the traffic sent by each connection for a particular time period and updates the weight values in a local table for future scheduling. Since reservation is done by sending control messages over the reservation channel, all neighbor nodes tuned to the reservation channel can update the local table.

The measured amounts of collected information for the first and second embodiments may include various metrics related to the provision and carrying load of the node or neighbor node and are described in detail below. In at least one embodiment of the present invention, it may be assumed that different priorities may correspond to different bandwidth requirements, delay requests and packet loss rates and may be defined for each traffic class or flow. In both embodiments, we can differentiate according to other priorities. The main difference is that in the second embodiment, additional neighbor information distributed over control frames is used to calculate channel access times.

Measured quantities are derived from the RTS and CTS messages and include packet length, data channel rate, source MAC address, destination MAC address, channel used (ie CTS, RTS, broadcast), packet priority level and weight. Contains information about the length of the queue being queued, but is not limited to this. The destination MAC address points to the next hop node address, and the packet priority level can be a combined value calculated from priority, backoff counter, age, etc., can be optional and used by the receiving node to occupy the current request. have. The weighted queue length may be a weighted value according to priority levels and traffic / stream indices to inform queue loads to be sent at later times. Rating may be done in terms of traffic classes or flows.

In a first embodiment of the scheduling algorithm, each node eavesdropping on the RTS and CTS message exchanges has a channel holdoff time based on packet length and data channel rate, and an address holdoff time based on source and destination address busy time. We will update the local traffic table by calculating

In a second embodiment of the scheduling algorithm, each node eavesdropping on RTS and CTS message exchanges may have a channel holdoff time based on packet length and data channel rate, an address holdoff time based on source and destination address busy time, And update the local traffic table by calculating the weight values.

Based on this information, in both the first and second embodiments, each node calculates a release time for the corresponding address and channels. This release time depends on the packet duration and the random time interval, which in turn depends on the node weight. The random interval may be, for example, a uniformly distributed time interval minus normalized node weight that may be standardized to have a direct proportional time intervals. Packet weight depends on priority, number of transmissions, age, and the like. The maximum value of the node weight depends on the individual packet weight values, buffer size and rate limiting algorithms used for each class or flow.

Each node finishing the transmission attempt may schedule the next attempt by setting these channel and destination release times with the calculated values. The packet will be sent if the destination and at least one channel are used. The timers are evenly distributed over several ranges depending on node weight, weighted previous packet transmission time and previous transmission status, defect or success.

In stateless scheduling, the first scheduling method, node weights are calculated as described for monitoring for the first and second embodiment modes of the scheduling algorithm. In stateful scheduling, the second scheduling method, when a node wishes to schedule its access to the first and second embodiment modes of the scheduling algorithm, its rank in the local table to calculate the channel access timer. Will check. Only when the channel access time is reached, and when the other node does not grab the channel, the node sends an RTS message to race the channel. If there are other transmission attempts during this channel access time, the node will reset the channel access timer and schedule a new channel access based on the new information.

For both the first and second embodiments of the scheduling algorithm, if the previous transmission attempt failed due to an incomplete RTS-CTS handshake, the channel access time will be increased to reduce the likelihood of collision. The increase can be exponential. If the previous transmission attempt failed due to negative arc or arc timeout, the node weight can be increased to compensate for channel damages.

One function of scheduling all of the above described algorithms is to decrease the channel access timer when the node's weight becomes large and to increase the channel access timer in the remaining situations. Ideally, the system should be fair for nodes with traffic of the same priority even at other provided loads. This can be accomplished with the calculation of maximum and minimum timer values, neighbor node aggregation traffic conditions and rate limiting algorithms.

For the first and second embodiments of the scheduling algorithm, when executing the second scheduling method, stateful scheduling, the algorithms include a function for tracking the carrying load of other nodes.

In particular, in a first situation in which a node receives an RTS message, regardless of whether an RTS message is intended for the node, the algorithm may provide an address busy time based on packet length and data rate, and provided load at certain priority levels. Calculate If the destination is not a neighbor, this indicates that the source that is not the destination is within the node's communication range. In both cases, the information in the RTS message, such as packet length, source, destination, and priority, is used to measure the traffic load provided.

In a second situation, when a node receives a CTS message that is not for the node, the algorithm receives the address and channel busy time, and the load carried, based on the packet length and the data rate at certain priority levels. .

In a third situation, when the node sends an RTS message, the algorithm calculates the provided load of the nodes at a particular priority level.

Finally, in the fourth situation, when the node receives the CTS message, the algorithm calculates the transmitted load of the nodes at a particular priority level.

If the receiving node detects that the transmitting node's transmitted load exceeds the allowed value, after receiving the RTS requesting the transmission, the receiving node sends a tuning factor to slow down the transmitting node by sending a tuning factor. You can decide to send a message. This eliminates many of the problems associated with hidden terminals, such as source-related unfair channels and destination access, and problems related to mobility, such as source-related unfair access due to lack of information about neighbors as the source moves to a new neighbor. The transmitter can adjust the new channel access time by using receiving node conditioning.

In addition, the first and second embodiments of the scheduling algorithm check the priority level and the weighted queue length to allow nodes in the same neighbor to schedule channel access to account for service discrimination and fairness. Priority and weighted queue lengths can adjust channel access delay by providing high priority for nodes with high priority and large weighted queue lengths, and the priority level packet age factor provides low priority traffic and small It can be used to adjust the channel access delay to prevent starvation of nodes with traffic load.

Embodiments of the present invention described above disclose a distributed self-coordinating approach for providing fairness and service differentiation in an ad hoc network by using control RTS / CTS messages carrying node state information. The local scheduling algorithm described above uses eavesdropping RTS / CTS information with attributes indicative of the states of the nodes in order to maximize the recognition of neighboring states for the multi-channel system, where data and Ack messages are transmitted over other data channels. Both the first and second embodiments of the scheduling algorithm measure the provided load, transmitted load, and reserved load in the neighborhood to adjust the channel access timer to provide fairness among other nodes, other links, and other streams. do. Each algorithm calculates both a priority level based on node queue state and a priority level based on neighbor state to destroy competition rankings, differentiate services, and avoid starvation of loads with low traffic loads. The receiving node is used to coordinate distributed scheduling decisions based on a better view of the receiving node relative to the neighbor.

The following is an illustrative description of an embodiment of the invention with embodiments. The following embodiments show the first and second algorithms where the node (source or relay node) has a set of traffic classes c. In these embodiments, the fairness interval is calculated after each transmission attempt, while channel and address monitoring is done after the receipt of an unintended RTS or CTS (ie RTS or CTS is not intended for this node). The above embodiments cover the transmission of unicast messages. Note that the calculation of backoff values, priority level and measurements may be changed and the above embodiments are provided for clarity.

First algorithm

Packet transmission:

1) The total "weight" is calculated by summing the priority values of the packets in the packet store. The priority of a packet is calculated as the weighted sum of the priority, the number of attempts and the age of the packets.

Where w _j is the total weight of node j, p _i is the priority level of the i th packet, and wp _c , wn _c and wa _c are the priorities (pri _i ) in the queue assigned for class (c), respectively. Weight factors for, number of attempts (na _i ) and age (a _i ) of the packet. Note that the weight corresponds to the provided rod. Although the relationship between node weight and priority information is linear in a given example, other types of relationships may be used, such as exponential or logarithmic. Similarly, various modifications and variations of the present invention, such as fairness interval calculations, may be made with these embodiments.

2) If a node is ready to send a packet with a new route to its destination, the next next hop waits for an RTS containing the set of available data channels, transmission period, source and next hop addresses. send. Note that because this is a multi-hop system, the new route must be one that can be used before transmission. Any change in the route can change the packet to be transmitted.

3) If the CTS times out or NCTS is received, update the fairness interval Tf.

Where wf is a positive number (if exponential backoff is used, wf is increased by 2 at every CTS receive coupling), Rf and Mf are positive numbers, and U [Rf] represents a uniform dispersion. Wn _j is a weight normalized to Mf-Wn _j ≧ n, where n is a positive number. For example, for n = 0, Wn _j = (W _j Mf) / W _max , where W _max is the maximum weight value.

4) If CTS is received, reduce wf or set wf = 1. The CTS includes the selected data channel, transmission period, source and next hop address. Check if the receiver has sent a new wf for this node. Note that the adjustment factor may include other parameters used in the backoff formula.

5) If Ack is received, update the fairness interval.

Where wt is a weight factor for the packet transmission period (pk_time). If someone wants to provide fairness in sharing relay nodes, the fairness interval for use of the next hop can be calculated similar to the channel fairness interval.

Address and Channel Monitoring:

1) After receiving the unintended RTS, mark the source and next hop addresses that have waited for the CTS and have been busy for the time interval Ta.

Where pk_time is the transmission period that includes the corresponding overhead due to the handshake technique. Rf and Mf may be different for fairness calculation and monitoring.

2) Mark the busy channel for source and next hop addresses and time interval Ta after receipt of the unintended CTS.

Where Rf and Mf may be different for address and channel monitoring.

Second algorithm

Transmission of packets

1) The total "weight" is calculated as the sum of the priority values of the packets in the packet store. The priority of a packet can be calculated as the priority, the number of attempts and the age of the packets. Note that every node distributes the weight value through the RTS and CTS messages.

Wherein W _j is the total weight of the node j, p _i is the priority level of a packet, wp _c, wn _c, and wa _c is the priority of the packet in each queue (C) (pr _i), may attempt (na _i ) and weight factors for the age (a _i ) of the packet and Wt _j is the node weight normalized according to the neighbor node weight.

2) If the node is ready to send a packet with a new route, the idle next hop includes a set of available data channels, transmission period, source and next hop address, node weight and channel access / throughput information. Send RTS.

3) If the CTS times out or NCTS is received, update the fairness interval Tf.

Where wf is a positive number (if exponential backoff is used, wf is increased by 2 for every CTS reception fault), Rf and Mf are positive numbers, and U [Rf] represents a uniform dispersion. Wtn _j is a weight normalized to Mf-Wtn _j ≧ n where n is a positive number. Note that the ranges for the parameters must be different than the first algorithm. Identical symbols are used for simplicity.

4) If CTS is received, wf is decreased or wf = 1 is set. Check if the receiver has sent a new wf value for this node. It is noted that the adjustment factor may include other parameters used in the backoff formulas.

5) If Ack is received, update the fairness interval.

Where wt is a weight factor for the packet transmission period (pk_time).

6) After receiving the ack, update the node channel occupation time. Other parameters such as throughput may be used.

Where α is the loss factor and l is the last time CT _j is updated. It is noted that α may depend on mobility. For example, α may be selected depending on the rate of change of the neighboring table. If someone wants to provide fairness in sharing relay nodes, it is noted that fairness for use of the next hop can be calculated similar to the channel fairness interval.

Address and Channel Monitoring:

1) After receipt of an unintended RTS, it waits for the CTS and marks both the source and next hop address vacated for the time interval Ta.

Where pk_time is the transmission period that includes the corresponding eavesdropping due to the handshake technique. Rf and Mf may be different from fairness calculation and monitoring. Update W _k and CT _k for this node. Compare the CT _k and W _k values with the CT _j and W _j values to update the Wf values. It is noted that the weight corresponds to the provided load and the channel occupancy information corresponds to the carried load.

2) After receipt of the unintended CTS, mark both the source and next hop addresses that have been busy for the time interval Ta. If not updated for RTS, update the W _k and CT _k values for this node.

Where Rf and Mf may be different for address and channel monitoring. The CT _k and W _k values are compared with the CT _j and W _j values to update the Wf values.

Although some embodiments of the invention have been described in detail above, those skilled in the art will readily recognize that many variations are possible in the exemplary embodiments without departing significantly from the new guidelines and advantages of the invention. Accordingly, all such variations are within the scope of the present invention.

Claims (7)

A method of performing medium access control (MAC) scheduling for nodes in an ad hoc multiple hopping peer-to-peer wireless communication network, the method comprising: In order to evaluate the priority level weight of the node based on the information contained in the received request to send (RTS) and clear to send (CTS) messages, Controlling the attempted node; Scheduling transmission of the node based on the evaluated priority level weights; The information contained in the RTS and CTS messages includes a weighted queue length calculated from source and destination MAC addresses, current packet length, data channel rate, selected data channel, and packet priority levels. and at least one of weighted queue lengths. The method of claim 1, The controlling further includes evaluating priority level weights of neighboring nodes based on the received RTS and CTS messages, And the scheduling step schedules the transmission based on the evaluated priority level weights of the node and the evaluated priority level weights of the neighboring nodes. delete A method of performing medium access control (MAC) scheduling for nodes in an ad hoc multiple hopping peer-to-peer wireless communication network, the method comprising: In order to evaluate the priority level weight of the node based on the information contained in the received request to send (RTS) and clear to send (CTS) messages, Controlling the attempted node; Scheduling transmission of the node based on the evaluated priority level weights; The information contained in the RTS and CTS messages may be carried in data and acknowledgment messages, respectively, if the system uses a single channel and RTS / CTS is not enabled for the transmission. A method of performing medium access control (MAC) scheduling for nodes in an ad hoc multiple hopping peer-to-peer wireless communication network, the method comprising: Attempt to transmit to evaluate a node's priority level weight based on information contained in received request to send (RTS) and clear to send (CTS) messages. Controlling a node to perform and evaluating priority level weights of neighboring nodes based on the received RTS and CTS messages; And  Scheduling transmission of the node based on the evaluated priority level weights of the node and the evaluated priority level weights of the neighboring nodes; The information contained in the RTS and CTS messages may be carried in data and acknowledgment messages, respectively, if the system uses a single channel and RTS / CTS is not enabled for the transmission. . The method of claim 1, And the information contained in the RTS and CTS messages may be carried in data and ack messages, respectively, if the system uses a single channel and RTS / CTS is not enabled for the transmission. The method of claim 1, And said node is a mobile node.

Images